Covariance structure and proc MIXED

[deleted_user]

Hi,

Could someone point me to a reference as to how SAS handles the computation of the different covariance matrices that can be used in proc MIXED for both random effects (RANDOM) and residuals (REPEATED)? I've been looking through the help file without being able to find anything.

It can either be an article in the help file or a section of a book: I don't mind.

Thanks a lot!

P.S. I use SAS v9.1 .

[1162]

It's been a while since I've done a repeated measures analysis, but I suspect you have to specify the covariance matrix in your RANDOM statement. Try looking at the TYPE= option.

Maybe the UN@AR(1) structure?

[SteveDenham]

Are you asking for the functional forms of the covariance matrices? If so, then look at the TYPE= portion of the documentation for the REPEATED statement.

"For further details about the computational techniques used in PROC MIXED, see Wolfinger, Tobias, and Sall (1994)." in the Computational Issues part of the Details: MIXED Procedure is a quote if you are asking about specific numerical methods. The full reference is:

Wolfinger, R. D., Tobias, R. D., and Sall, J. (1994), "Computing Gaussian Likelihoods and Their Derivatives for General Linear Mixed Models," SIAM Journal on Scientific Computing, 15(6), 1294–1310.